一种新的基于网格编码和区域合并的SAR图像快速分割算法  

A New Fast SAR Image Segmentation Algorithm Based on Grid Coding and Region Merging

作　　者：张泽均[1] 水鹏朗[1] 

机构地区：[1]西安电子科技大学雷达信号处理国家重点实验室,西安710071

出　　处：《电子与信息学报》2014年第4期974-980,共7页Journal of Electronics & Information Technology

基　　金：国家自然科学基金(61271295)资助课题

摘　　要：该文建立一种新的基于八邻域网格编码的SAR图像分割模型,并用区域合并技术实现了模型的快速求解。利用多方向比例边缘检测算子提取SAR图像的比例边缘强度映射(RESM),提出一种新的阈值处理方法抑制RESM均质区域内部的极小值,进而减少了对阈值处理后的RESM进行分水岭变换获得的初始分割的区域个数。递归地合并相邻区域来求取分割模型的次优解。利用区域邻接图(RAG)及其最近邻图(NNG)特性来加速区域合并过程。引入精确度(P)和召回率(R)来评价分割算法的边缘定位精度。与常用方法相比,该文方法具有高的边缘定位精度和低的时间复杂度。A new SAR image partition model is constructed based on 8-neighbor grid code, which is fast solved by region merging. Utilizing multi-direction ratio edge detector to construct Ratio Edge Strength Map （RESM） of SAR image, a novel thresholding method is proposed to suppress the minima value in the homogeneous region of RESM, which reduces the number of regions in an initial partition produced by watershed of the thresholding processed RESM. Sub-optimization of the partition model is obtained by merging adjacent region pair iteratively. Region Adjacency Graph （RAG） and its Nearest Neighbor Graph （NNG） characteristic are used to speed up the proceeding of region merging. Precision （P ） and Recall （R） are introduced to evaluate the boundary localization precision of segmentation methods. Compared with three widely used methods, the proposed method has higher boundary localization precision and lower computational complexity.

关 键 词：SAR图像分割 网格编码 快速区域合并 区域邻接图(RAG) 最近邻图(NNG) 

分 类 号：TP751.1[自动化与计算机技术—检测技术与自动化装置]